This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DAC3282: DAC3482 Spur Issue

Part Number: DAC3282
Other Parts Discussed in Thread: LMK01000, DAC3482

Hi,

 

I am running DAC3482 from my own board. An xilinx K7 FPGA is used to drive DAC through LVDS and controll onboard DDR memory. An external 1228.8MHz clock source is applied to LMK01000 then distributed to DAC and FPGA.

DAC CLK = 1228.8MHz

Interpolation = x4

16 bit Data Mode

2's Component Disabled [data range 0 ~ 65535]

DAC IF freq = DAC CLK/16 = 76.805MHz

 

DAC was configured by MSP430 MCU as below:

DAC3482_RST_LOW;

delay_ms(10);
DAC3482_RST_HIGH;

DAC3482_TXEN_LOW;
unsigned int cfg2 = 0;
if(Dac3482_Twos) cfg2|= BIT1;//2's disabled
//if(Dac3482_xIF) cfg2|= BIT4;

Dac3482WriteSpi(0x0280D0 | cfg2); 

unsigned int cfg0 = Dac3482_xINT>>1;//Dac3482_xINT = 4
cfg0 = cfg0 <<8;
cfg0 = cfg0 + 0x1C + BIT7; 
Dac3482WriteSpi(cfg0); //Reg 00 

Dac3482WriteSpi(0x01020E); //Reg 01
Dac3482WriteSpi(0x03A000); //Reg 03 
Dac3482WriteSpi(0x103000); //Reg 10 

Dac3482WriteSpi(0x140000); //Reg 14 - 15:0

unsigned long cfg15 = 0x150000;
if(Dac3482_xIF>1)
cfg15 = cfg15 + 0x10000/Dac3482_xIF;//Dac3482_xIF = 16
Dac3482WriteSpi(cfg15); //Reg 15 - 31:16 

Dac3482WriteSpi(0x1B0800); //Reg 1B 
Dac3482WriteSpi(0x1F8180); //Reg 1F 
Dac3482WriteSpi(0x202201); //Reg 20

//Frame 
//FPGA RST
FPGA_RST_LOW; 
delay_ms(100);//apply Frame 

Dac3482WriteSpi(0x050000); //Reg 5
Dac3482WriteSpi(0x1F818A); //Reg 1F
Dac3482WriteSpi(cfg0); //Reg 0
Dac3482WriteSpi(0x1F8188); //Reg 1F
Dac3482WriteSpi(0x200001); //Reg 20

delay_ms(100); 
DAC3482_TXEN_HIGH;

Here are some test data,

DAC IF Output (no iq data)

 

play a 48-point-sine wave (data file contains 819200 cycles)

data frequency = dac clk / 48 / 4 = 6.4MHz; Output freq = 76.805 + 6.4 = 83.205M


play 20M LTE waveform( 307.2M sample rate)


The LTE waveform data file spectrum is clean in Matlab; the data file is also verified by HSDC+TSW1400+TSW3085. So waveform file cann't be the reason.


 

I don't understand why threre so may spurs when playing the modulation data while everything is ok with sine wave. 

Is there anybody can help to remove those spurs?

 

Any idea will be highly appreciated.

Thanks in Advance!

TL

 

 

 

  • Hi TL,

    We are taking a closer look into your question, and will be back with you soon.

    Are you able to share a schematic of your DAC? It may be helpful to see the output stage of the DAC.

    Best Regards,

    Dan

  • TL,

    when such modulation occurs with the LTE, it is due to the LTE signal being cut-off abruptly. You will need to check your FPGA DDR playback whether it is multiples of N samples. The LTE waveform generated from the TSW1400 LTE tool are usually non-integer multiples of N (i.e. multiple of 8 or 16, or 32). This causes the playback at the end to cut-off suddenly, and the signal is no longer coherent. 

    The fact that you verified the waveform + TSW3085 are fine would support this theory. It is possible that your FPGA setup requires multiple of N DDR readback.

    -Kang